Skip to main content
ARS Home » Pacific West Area » Corvallis, Oregon » Forage Seed and Cereal Research Unit » Research » Publications at this Location » Publication #420371
Research Project: Improving Plant, Soil, and Cropping Systems Health and Productivity through Advanced Integration of Comprehensive Management Practices

Location: Forage Seed and Cereal Research Unit

Title: What is the (real) rate of soil health practice adoption? Making sense of three data sources

Author
item MCGILL, BONNIE - American Farmland Trust
item PUNTEL, LAILA - Syngenta
item HIVELY, DEAN - Us Geological Survey (USGS)
item SHRIVER, JOHN - Iowa State University
item Thieme, Alison
item Manter, Daniel
item Moore, Jennifer

Submitted to: Journal of Soil and Water Conservation
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/3/2025
Publication Date: 1/27/2026
Citation: Mcgill, B., Puntel, L.A., Hively, D.W., Shriver, J., Thieme, A.N., Manter, D.K., Moore, J.M. 2026. What is the (real) rate of soil health practice adoption? Making sense of three data sources. Journal of Soil and Water Conservation. https://doi.org/10.1080/00224561.2025.2580218.
DOI: https://doi.org/10.1080/00224561.2025.2580218

Interpretive Summary: Conservation stakeholders looking to quantify the impact of soil health practices on croplands often face challenges in interpreting practice adoption data due to inconsistencies among data sources. These data can be derived from various methodologies such as surveys, on-site visits, or remote sensing data. In this paper, we compared data from each of these methodologies and identified some discrepancies which were largely attributable to differences in scope (i.e., assessment of practice implementation vs efficacy) which will provide users with more clarity on why the methods may vary and increase our ability to make evidence-based decisions on the role and impact of conservation practice adoption in agricultural systems.

Technical Abstract: Conservation stakeholders looking to quantify the impact of soil health practices on croplands often face challenges in interpreting practice adoption data due to inconsistencies among data sources. Similarly, environmental modeling efforts can yield different outputs based on their estimated rates of practice adoption. To help make sense of different adoption data sources, we compared county-level adoption data for winter cover crops (CC), no-till (NT), and reduced tillage (RT) in three areas of the United States: Central Illinois (CIL), Southern Illinois (SIL), and Western New York (WNY). We analyzed data available for 2015 to 2022 from the Operational Tillage Information System (OpTIS, based on remote sensing), US Census of Agriculture (based on farmer response to a survey), and, specifically in Illinois, the Illinois Soil Conservation Transect Survey (a roadside survey). The magnitude of differences between the datasets was dependent upon the practice and geographic location. For example, there were more instances of agreement between OpTIS and Transect tillage data (2015, 2017, and 2018) in CIL compared to SIL. Averaged across regions in Illinois, differences between 2017 OpTIS and AgCensus tillage data were relatively small (less than 2.5 percentage points on average), but up to an average of 35 percentage points in WNY. There was less variability and smaller differences between OpTIS and AgCensus cover crop data in Illinois compared to WNY. AgCensus tended to report lower cover crop adoption for Illinois and greater adoption in WNY compared to OpTIS. Differences between the datasets were attributed to definitional inconsistencies for RT and NT, and how cover crop data were acquired. For example, the census asks if a cover crop was planted, regardless of establishment success, whereas OpTIS and Transect evaluate the presence of a standing cover crop. Data sources also differed by the type of cropland assessed (corn, soybean, or all cropland). We propose three recommendations to improve interpretation and consistency. (1) Recognize the scope of each data source, what type of adoption data (implementation or efficacy) are needed for making a specific conservation decision and use the most appropriate data. (2) Initiate a consensus-building process to harmonize definitions and methodologies. (3) Develop a public, integrated survey, transect, and remote sensing practice adoption data program at field and national scales. Such a research effort would improve data access and utility for evidence-based conservation decision-making by organizations and enhance the accuracy of environmental models that rely on adoption data as input.